Given all these drawbacks to the rate-of-returns approach, it is not surprising that governments in developing countries typically ignore this approach to educational planning and instead adopt a "manpower approach," attempting to tailor educational development to projected manpower requirements.* Nor is it surprising that, although the economic benefits of lowering the birth rate have been recognized as a valid reason to invest in family planning programs, strict rate-of-return analysis has often been given a wide berth.t But to question our ability to measure accurately the rate of return on either education or family planning and to doubt the appropriateness of relying on the rate-of-returns approach alone are not to deny the place of strict rate-of-returns analysis in a broader cost-benefit approach to the problem of allocating government investment. Such an approach should also take into account welfare criteria, political considerations, and ideally a variety of spinoffs that cannot be measured very precisely, such as the modernizing effect of family planning programs and formal education on people's conception of their ability to influence their own destiny.
The current increase in expenditure on both education and family planning is in itself evidence of the need for a more careful evaluation of the costs and benefits of these expenditures in the future. As noted earlier, one cannot confidently set an upper limit to the possible level of educational expenditures, and a few developing countries have already lifted their spending on education above 7 percent of GNP. A sharp increase in the share of budgets going into education is no doubt difficult to achieve in any country, but the potential for such increases, given a reordering of national priorities, is clear enough. For example, education's toughest competitor for funds in many countries is the military budget, which in Middle Eastern countries averages more than 11 percent of GNP (slightly higher than in the U.S.S.R. and the United States), and in the rest of Asia averages more than 6 percent of GNP (both figures medians).?
Given that education's share of the total budget can, and probably will, be raised in most countries, it becomes urgent that the question be faced: Where is the point at which educational expenditure becomes excessive? Should
"This approach is not necessarily any more useful. An incisive critique is contained in (71, pp. 3-18). See also (80).
t Recognition of the economic benefits of investing in family planning, of course, is not dependent upon acceptance of the cost-benefit approach outlined above. Given the assumption that family planning programs have an independent effect on the birth rate by aiding people to prevent unwanted births (and thus by definition increasing individual satisfaction), the economic benefits arc demonstrated by growth models of the kind cited in (56). Indeed, Simon has derived a figure for the value of a prevented birth from such a model (78).
$In the giant countries of the region-India, Pakistan, and Indoncsia-howevcr, a smaller than average share of GNP goes to defense-3.5 percent, 3.9 percent, and 2.5